[Bug] staking validator-history fails: "client.getSlashingAddress is not a function"

[Tobu8888]

Description

genlayer staking validator-history throws client.getSlashingAddress is not a function and fails to return any history. All other staking subcommands work fine on the same account/network, so the bug appears isolated to the validator-history code path.

Steps to reproduce

1. Configure the CLI for testnet-asimov with a valid account.
2. Run:

   genlayer staking validator-history --validator 0x9f50DDcd4Af137E14467232490035F13D67998C0

Actual result

⠸ Fetching validator history...
Error:
 client.getSlashingAddress is not a function
✖ Failed to get validator history

Expected result

Returns the validator's slash/reward history (default: last 10 epochs).

Works fine on the same account/network (isolates the bug)

validator-info, epoch-info, active-validators, set-operator, and set-identity all run successfully — only validator-history fails.

Environment

• genlayer-cli: 0.39.1
• Network: testnet-asimov (Asimov Phase 5)
• OS: Linux x86_64
• Node.js: v20.20.2

Additional notes

The error suggests the client object is missing a getSlashingAddress method — possibly renamed/removed in the underlying SDK, or not wired up for the testnet-asimov network in the validator-history path.

[Jr-kenny]

Dug into this. The root cause is the dependency pin, not the command logic.

genlayer-cli pins genlayer-js#v2-dev, which is a moving branch. validator-history was written when v2-dev still had client.getSlashingAddress(), which read the staking contract's contracts() tuple and returned the slashing address at index 4. genlayer-js #131 (commit ad573bb, Feb 10) updated the staking ABI to the v0.5 IGenLayerStaking Diamond interface and removed the method because contracts() no longer exists in that interface. Since the dependency resolves the branch tip at install time, every fresh install of the released CLI gets a validator-history that calls a method that no longer exists, which is exactly the TypeError above. The other staking subcommands keep working because they never touch the removed method, matching what you observed.

On the fix: v0.5 is a Diamond interface, so the slashing facet emits SlashEnacted and SlashedFromIdleness from the diamond address itself. That means validator-history can fetch the slash logs from the staking contract address and drop the second address lookup entirely. The SLASH_ABI events are still exported by genlayer-js, only the address discovery died. The same moving-branch pin will keep producing this class of breakage, so pinning to a commit might be worth a separate conversation.

Happy to take this and put up a PR if you assign me.